Metraipolis › archive › dc_9bbae9e29a08

In RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Post-Retrieval Context Tampering — arXiv:2609.09243

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.09243
License (as archived)
CC0-1.0
Archived
2026-10-16T07:25:07.727Z
SHA-256 of the archived text
c8128fe5c368527b13c6ab2804c60f0cdbbe7be0bb7ac0af0326e2f690ef0a99
Stable URL of this page
https://metraipolis.com/documento/dc_9bbae9e29a08

How to cite

In RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Post-Retrieval Context Tampering — arXiv:2609.09243 — arXiv (en), https://arxiv.org/abs/2609.09243 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9bbae9e29a08 (Metraipolis archive dc_9bbae9e29a08, sha256 c8128fe5c368527b13c6ab2804c60f0cdbbe7be0bb7ac0af0326e2f690ef0a99)
BibTeX
@misc{dc9bbae9e29a08,
  title = {In RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Post-Retrieval Context Tampering — arXiv:2609.09243},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.09243},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9bbae9e29a08; sha256 c8128fe5c368527b13c6ab2804c60f0cdbbe7be0bb7ac0af0326e2f690ef0a99},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09243. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieval-augmented generation (RAG) grounds a language model in retrieved documents, which reduces hallucination but creates a new attack surface: if retrieved text is tampered with, the model may repeat the falsehood. We study how much a small quantized model, Llama 3.1 8B, degrades when a fraction of its retrieved context is poisoned. Three corruption strategies are tested, entity swap, number swap, and negation, each applied to zero, one, two, or three of the three retrieved passages, over a factorial sweep of 588 runs on a fact-checking task built from FEVER. Accuracy falls from 77.9% on clean context to 43.5% when all three passages are corrupted. Entity swap flips the largest share of answers that were correct on clean context. Number-based corruption stays flat while poisoned passages are a minority and jumps once they form a majority, a pattern we re-check with query-level bootstrap intervals. The model rarely invents new falsehoods; its dominant reaction is to abstain, and a lexical overlap proxy of unsupported generation falls under attack rather than rising. The study is a small-scale measurement with coarse automated labels; we treat the strategy contrasts as suggestive until decoding is controlled and stronger adjudication is in place.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-10 in cs.AI).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.09243 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T07:00:16.286Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).